October 2015

IZA DP No. 9452: A Longitudinal Analysis of Violence and Housing Insecurity

published in: Economic Inquiry, 2018, 56(3), 1602-21.

Violence and housing insecurity are horrible events that may be intertwined, with violence possibly forcing victims to abandon their accommodations and housing insecurity depriving people of the safety of a home or placing them in compromised circumstances. This study uses national, prospective, longitudinal data from the Journeys Home Survey to examine how violence, housing insecurity, and other characteristics in one period affect disadvantaged Australian men's and women's chances of experiencing violence and housing insecurity in subsequent periods. The study is one of the first to investigate these relationships prospectively and unusual in considering how violence among adult men contributes to their housing insecurity. We estimate dynamic multivariate models that control for observed and time-invariant unobserved characteristics and find that men's chances of being housing secure without experiencing violence are 24-45 percent lower and women's chances are 12- 20 percent lower if they experienced housing insecurity, violence or both in the previous period. Heavy drinking, marijuana use, psychological distress, and a history of childhood abuse and neglect also increase the risks of violence and housing insecurity for both genders, while the presence of children reduces these risks. Women who are bisexual or lesbian and women with homeless friends also face elevated risks of housing insecurity, while men's sexual orientation and friend networks seem less relevant.